IRC Logs for #crux-devel Tuesday, 2010-03-09

*** Rotwang has joined #crux-devel03:02
*** tnut has left #crux-devel03:31
*** Rotwang has quit IRC04:26
*** tnut has joined #crux-devel04:33
*** sepen has joined #crux-devel05:30
*** tnut has quit IRC09:03
jaegertesting the USB image install in a VM09:40
jaegerinstalled without errors, yay09:48
*** pitillo has quit IRC09:51
*** pitillo has joined #crux-devel10:12
sepennice! but I can't understand where is the problem about img creation10:17
jaegersepen: the problem is doing it with some useful variables in a Makefile, as well as fdisk scripted input, etc.10:19
jaegerI'll upload the script I made, sec10:19
sepenjaeger, didn't work this stuff? http://crux.nu/~sepen/usbdisk/10:20
jaegersepen: I'm sure it works fine but it isn't in the system/iso.git Makefile, which is what I was trying to do10:21
jaegerhttp://jaeger.morpheus.net/linux/crux/tmp/iso-patches/10:22
jaegerIt looks like the best we could do is to call a script from the Makefile10:22
jaegerif that's your make_usbdisk script, that's fine with me10:22
sepenI tried a patch for 2.5 in the past10:22
sepenbut if I remember correctly cptn told me that release the image don't make sense10:23
jaegerDo you remember why?10:23
sepenanother 270megs?10:24
jaegerIt doesn't have to be hosted on crux.nu if that's the problem, I guess10:24
sepenyep10:24
sepenjaeger, ah I've a patch at home useful for dynamic cpio images10:25
sepenso we can avoid the initramfs.lst file10:25
jaegerhow so? what does it do?10:25
sepenpipes the output to a command and creates the cpio archive10:25
sepens/to/from10:26
sepenjust to avoid the static initramfs.lst10:26
sepenI'll do more tests and report results here10:26
sepenbut I need to find some time10:27
sepenbbl10:27
jaegerIf you remove the initramfs.lst file how does it ignore modules you might not want in the initramfs?10:27
*** sepen has quit IRC10:56
*** tnut has joined #crux-devel11:38
*** tnut has quit IRC12:54
*** tnut has joined #crux-devel13:08
*** tnut has quit IRC13:14
*** sepen has joined #crux-devel13:47
sepenjaeger, I think we don't need fdisk for the script13:48
sepenso you can use directly sda instead of sda1, etc.13:48
sepenand yes, without initramfs.lst I think you can create the initramfs image by doing $ cd $PATH; find . | cpio -o -H newc | gzip > $INITRAMFS_FILE13:55
sepenthat will make the initramfs more hack'able imho13:56
jaegerwhat if you want to leave specific modules out? you'd have to do it from the kernel config itself, I guess14:10
sepenjust move those dir modules out of the $PATH14:22
jaegerI'm not seeing any actual benefit here, what am I missing?14:27
sepenit could be more maintainable imho and wouldn't be necessary to check for new modules when kernel config grows up14:34
sepenjaeger, I do not try to convince you, just that is a idea ;D14:38
jaegerI'm not against it at all if it's better, just curious14:39
sepenwell the fact is that I read that from the kernel doc as another way to create ram filesystems14:39
sepenfrom ramfs-rootfs-initramfs.txt14:40
jaegeryeah, I've read some of that stuff as well, just wondered14:42
sepentoo many useful, I used it to boot my old laptop from floppies :D14:46
teK_anyone with 32bit willing to compile cvs?14:47
teK_won't work for me with 64 Bit14:48
jaegerteK_: tried it with linux32?14:49
teK_I have no 32 bit system available14:49
jaegerno multilib I take it14:50
teK_I'm running pure 64 bit14:50
jaegerah14:50
teK_only drawback is that I don't have wine but well.. I do not really need it14:51
jaegeryou could always convert to multilib if you did need it14:52
teK_there's no package ;(14:52
jaegergive me a min, I'll build the 32-bit package14:52
teK_thanks14:52
jaegeruse the 32-bit one :)14:52
teK_I don't want 32bit tbh14:52
jaegerok, uploaded a package to http://jaeger.morpheus.net/misc/14:57
teK_so it works for 32bit14:59
teK_okay, thanks14:59
jaegernp15:03
*** Rotwang has joined #crux-devel15:22
*** Rotwang has quit IRC20:04
*** sepen has quit IRC20:20

Generated by irclog2html.py 2.11.0 by Marius Gedminas - find it at mg.pov.lt!